Search

Discovery studies in animals constitute a cornerstone of biomedical research, but suffer from lack of generalizability to human populations. We propose that large-scale interrogation of these data could reveal patterns of animal use that could narrow the translational divide. We describe a text-mining approach that extracts translationally useful data from PubMed abstracts. These comprise six modules: species, model, genes, interventions/disease modifiers, overall outcome and functional outcome measures. Existing National Library of Medicine natural language processing tools (SemRep, GNormPlus and the Chemical annotator) underpin the program and are further augmented by various rules, term lists, and machine learning models. Evaluation of the program using a 98-abstract test set achieved F1 scores ranging from 0.75-0.95 across all modules, and exceeded F1 scores obtained from comparable baseline programs. Next, the program was applied to a larger 14,481 abstract data set (2008-2017). Expected and previously identified patterns of species and model use for the field were obtained. As previously noted, the majority of studies reported promising outcomes. Longitudinal patterns of intervention type or gene mentions were demonstrated, and patterns of animal model use characteristic of the Parkinson's disease field were confirmed. The primary function of the program is to overcome low external validity of animal model systems by aggregating evidence across a diversity of models that capture different aspects of a multifaceted cellular process. Some aspects of the tool are generalizable, whereas others are field-specific. In the initial version presented here, we demonstrate proof of concept within a single disease area, Parkinson's disease. However, the program can be expanded in modular fashion to support a wider range of neurodegenerative diseases.

We have previously described a user-interactive rule-based computer program (Dyna-SaurI) designed for dynamic thesaurus integration, and demonstrated its efficacy on integrating dermatological subsets of the MeSH and SNOMED thesauruses. In the present study, we have refined our rules for merging and mapping multiple thesauruses and tested these rules. We then applied them with a set of optimized parameters to the integration of a third thesaurus, a subset of the International Coding Index for Dermatology, with the Integrated MeSH-SNOMED thesaurus. The parameter changes resulted in improved ranking of more specific and conceptually closer terms.

MEDLINE® retrieval using several information retrieval algorithms was characterized for relevance to point-of-care therapeutic decisions for a sample of clinical queries in family practice. Evaluation methodology is described and preliminary results are presented.

Dermatologists were shown digitized images of skin lesions compressed by Joint Photographic Expert Group (JPEG) and fractal image format (FIF) algorithms and asked to make a diagnostic assessment and rate the image quality in displaying the features necessary for that assessment. Dermatologists performance in the diagnostic assessment was not affected by the level of compression (up to 40×) using either algorithm, but ratings of image quality for all compressed images were significantly lower than for uncompressed images and for the original photographic slides. There were no significant differences in dermatologists' performance or ratings between the two compression algorithms.

The research reported in this paper concerns an evaluation of the impact of compression on the quality of digitized color dermatologic images. 35 mm slides of four morphologic types of skin lesions were captured at 1000 pixels per inch (ppi) in 24 bit RGB color, to give an approximate 1K X 1K image. The discrete cosine transform (DCT) algorithm, was applied to the resulting image files to achieve compression ratios of about 7:1, 28:1, and 70:1. The original scans and the decompressed files were written to a 35 mm film recorder. Together with the original photo slides, the slides resulting from digital images were evaluated in a study of morphology recognition and image quality assessment. A panel of dermatologists was asked to identify the morphology depicted and to rate the image quality of each slide. The images were shown in a progression from highest level of compression to original photo slides. We conclude that the use of DCT file compression yields acceptable performance for skin lesion images since differences in morphology recognition performance do not correlate significantly with the use of original photos versus compressed versions. Additionally, image quality evaluation does not correlate significantly with level of compression.© (1991) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Color 35mm photographic slides are commonly used in dermatology for education, and patient records. An electronic storage and retrieval system for digitized slide images may offer some advantages such as preservation and random access. We have integrated a system based on a personal computer (PC) for digital imaging of 35mm slides that depict dermatologic conditions. Such systems require significant resources to accommodate the large image files involved. Methods to reduce storage requirements and access time through image compression are therefore of interest. This paper contains an evaluation of one such compression method that uses the Hadamard transform implemented on a PC-resident graphics processor. Image quality is assessed by determining the effect of compression on the performance of an image feature recognition task.© (1990) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

A microcomputer-based system for copying, storing, retrieving, and displaying color still images was assembled from commercially available components. The system was tested against 35-mm color photographic transparencies by measuring the performance of 91 medical students in a task of identifying the primary morphology of eight skin lesion images from the displayed images. There was no significant difference in the students' performance between the two display media.

THE INCIDENCE of and mortality from cervical cancer in the United States have decreased dramatically over the past 40 years, in part because of early diagnosis and treatment of cervical cancer precursor lesions. The success of cervical cytological screening has served as a model for early diagnosis of other types of cancer. Although numerous studies have shown that lack of cytological screening is a major risk factor for the development of invasive cervical cancer, 1-3 it is important to emphasize that none of the screening, diagnostic, or therapeutic techniques used in medicine are perfect. Accordingly, a few women will develop cervical cancer despite adherence to accepted screening protocols . In addition, problems inherent with sampling, interpretation, and effective clinical follow-up preclude total prevention of cervical cancer. In recent years it has become evident that the cost and morbidity associated with the detection and treatment of low-grade cervical lesions have escalated, probably

Automatic image annotation is an attractive approach for enabling convenient access to images found in a variety of documents. Since image captions and relevant discussions found in the text can be useful for summarizing the content of images, it is also possible that this text can be used to generate salient indexing terms. Unfortunately, this problem is generally domain-specific because indexing terms that are useful in one domain can be ineffective in others. Thus, we present a supervised machine learning approach to image annotation utilizing non-lexical features extracted from image-related text to select useful terms. We apply this approach to several subdomains of the biomedical sciences and show that we are able to reduce the number of ineffective indexing terms.